A Biodegradable Formulation for Lacewings to Trigger Biological Control of Aphids

Article Recruiting on the Spot: A Biodegradable Formulation for Lacewings to Trigger Biological Control of Aphids

Joakim Pålsson 1,†, Gunda Thöming 2,†, Rodrigo Silva 3, Mario Porcel 1,‡ , Teun Dekker 1 and Marco Tasin 1,* 1 Department of Plant Protection Biology, Swedish University of Agricultural Science, 230 53 Alnarp, Sweden; [email protected] (J.P.); [email protected] (M.P.); [email protected] (T.D.) 2 NIBIO, Norwegian Institute of Bioeconomy Research, Postbox 115, NO-1431 Ås, Norway; [email protected] 3 Isca Technologies Inc., 1230 Spring St., Riverside, CA 92507, USA; [email protected] * Correspondence: [email protected]; Tel.: +46-40-415320 † Shared First Authorship. ‡ Current address: Corporación Colombiana de Investigación Agropecuaria (Agrosavia), C. I. La Libertad, Vía Puerto López km 17, Meta, Colombia

Received: 28 November 2018; Accepted: 29 December 2018; Published: 5 January 2019

Abstract: Upon herbivory, plants release herbivore-induced plant volatiles (HIPVs), which induce chemical defenses in the plant as well as recruit natural enemies. However, whether synthetic HIPVs can be employed to enhance biological control in a cultivated crop in the field is yet to be explored. Here we show that a biodegradable formulation loaded with induced and food-signaling volatiles can selectively recruit the common green lacewing, Chrysoperla carnea, and reduce pest population under field conditions. In apple orchards, the new formulation attracted lacewing adults over a 4-week period, which correlated well with independent assessments of the longevity of the slow-release matrix measured through chemical analyses. In barley, lacewing eggs and larvae were significantly more abundant in treated plots, whereas a significant reduction of two aphid species was measured (98.9% and 93.6% of population reduction, for Sitobion avenae and Rhopalosiphum padi, respectively). Results show the potential for semiochemical-based targeted recruitment of lacewings to enhance biological control of aphids in a field setting. Further research should enhance selective recruitment by rewarding attracted natural enemies and by optimizing the application technique.

Keywords: apple; barley; Chrysoperla carnea; herbivory induced plant volatiles; integrated pest management; semiochemicals; ecological intensiﬁcation

1. Introduction Intensification of food production has been at odds with ecological sustainability. Among the unintended side effects are the loss of biodiversity, the simplification of the landscape, and an increased reliance on pesticides [1]. While these production systems favor high yields, corresponding repercussions on multitrophic relationships, such as pest’s natural enemy interactions, have been significant [2]. Ecological intensification is proposed as an approach to restore multifunctionality in agro-ecosystems, while supporting high productivity levels [3]. Enhancing biological control of insect pests through ecological processes is among the objectives of the “ecological paradigm shift” [3]. While an increasing body of evidence demonstrates how botanical diversity can effectively prevent pest attacks in crops by re-establishing trophic cascades [4], practical knowledge and cultural and economic gaps often impede the adoption of this concept in both high and low-input agricultural systems [5,6]. New methods should thus be developed to gradually transit from current input intensive monoculture to ecologically intensified crops.

Insects 2019, 10, 6; doi:10.3390/insects10010006 www.mdpi.com/journal/insects Insects 2019, 10, 6 2 of 15

Plants developed chemical and physical defenses to counter herbivory, some of which are only expressed upon attack. These inducible defense mechanisms can be particularly advantageous when herbivores are diverse and variable in time and space, and allow plants to optimize resource allocation and tailor the response to the current threats [7]. Among inducible defenses, herbivory-induced plant volatiles (HIPVs) rapidly prime surrounding tissue and neighboring plants and recruit natural enemies from a distance. In order to induce systemic defense in the whole plant, HIPVs are released from the damaged area into the atmosphere [8,9]. A faster induction at the individual plant level may be achieved through the atmosphere, especially in large plants such as trees. HIPVs additionally allow the releasing plant to communicate with other trophic levels, attracting predators and parasitoids [10]. As the emission of HIPVs occurs exclusively upon herbivory, they provide an honest cue to both specific and generalist third trophic level arthropods. Given such properties, synthetic HIPVs may be used as a tool for pest management in agriculture [11]. Several suggestions on how to utilize HIPVs in crops are reported in literature, either as a way of monitoring beneficial arthropods within a crop or to attract natural enemies in an attempt to directly control pest populations [12]. Indeed, HIPVs have been found to attract natural enemies belonging to different families, including hoverflies (Diptera: Syrphidae), predatory bugs (Heteroptera), ladybirds (Coleoptera: Coccinellidae), predatory mites (Mesostigmata), parasitic wasps (Hymenoptera), and green lacewings (Neuroptera: Chrysopidae) [11]. Because the composition of the volatile blend was found to fine-tune the range of attracted natural enemies, information on possible synergies among volatiles with a diverse ecological function could be used to specifically modulate behaviors of such beneficial arthropods [13]. For example, acetic acid (AA), which is often considered a sugar signaling compound, enhanced attraction of a number of insect species to plant volatiles [14,15]. Similarly, phenylacetaldehyde (PAA), which is a nectar-signaling floral as well as an induced volatile [16,17], was reported to increase predator attraction when added to other plant volatiles [16]. The synergy between food-signalling cues and classical inducible compounds, such as methyl salicylate (MS), which is attractive for several predatory groups [18], may be particularly interesting in species where biological control is not directly exerted by the attracted adults, but rather by their offspring. One such biological control agent is the common green lacewing, Chrysoperla carnea s.l., adults of which feed on nectar and pollen, while the larvae are generalist predators [19]. Several compounds attracted significant amounts of lacewings (i.e., Acetophenone, 2-phenylethanol (PE), PAA) in cherry, apricot, apple, pear, and walnut orchards [16,20,21]. Interestingly, a blend of MS, PAA, and AA not only attracted adults, but was also capable of increasing the oviposition rate in the vicinity of the releasing point, possibly thanks to the presence of food signalling volatiles [22]. In addition, this blend significantly increased lacewing density in overwintering boxes [22]. Whether this blend is capable of enhancing pest control in the surrounding vegetation remains to be investigated. We hypothesized that a novel formulation consisting of plant and food odors embedded in a biodegradable matrix can attract lacewing from surrounding vegetation and induce oviposition on the plant. We further hypothesized that this recruitment of adults would contribute to the suppression of aphid populations. Common green lacewing attraction from a distance and release of volatiles from emitting devices were evaluated in apple orchards (Malus domestica Borkhausen). However, the assessment of biological control was carried out in a cereal field, due to a higher and much more homogeneous pest attack in comparison to orchards. Parameters such as larval lacewing density and population level of other natural enemies were measured in the system of barley (Hordeum vulgare L.), two hemipteran herbivores, Sitobion avenae (F.) and Rhopalosiphum padi (L.), and the common green lacewing as a generalist predator. In order to propose a method with a low environmental impact, we compared the efficacy of a standard polyethylene-based bag odor dispenser with a newly assembled prototype based on a biodegradable matrix. Insects 2019, 10, 6 3 of 15

2. Materials and Methods

2.1. Volatile Releasing Formulations The reference device was purchased from Csalomon (Plant Protection Institute, MTA ATK, Budapest, Hungary). It consisted of a cotton wick loaded with a 3-component blend of MS, PAA, and AA in a 1:1:1 ratio with a total load of 300 mg/device. The wick was placed into a sealed polyethylene bag, through which volatiles were slowly released (hereafter referred to as PE bag). In order to facilitate its use in the ﬁeld, this device was delivered with a pre-attached plastic strip to be stapled into the canopy, or to a holding stick. We selected this formulation as a benchmark because ﬁeld data on lacewing attraction and oviposition using this formulation were available at the time of our study [20]. The new formulation was a novel product prepared in co-operation with ISCA Technologies (ISCA Technologies Inc., Riverside, CA, USA) and Bio-Innovate AB (Lund, Sweden). It consisted of a biologically inert, biodegradable wax-water emulsion releasing paste loaded with the above-described blend at a concentration of 300 mg/mL. A single release point for this product constituted a 1 mL droplet applied with a plastic syringe (hereafter referred to as paste). Data on the slow-release properties of the paste using other volatile cues are also available [23].

2.2. Measurement of Volatile Release To compare over time the volatile release rate from the two formulations, PE bag and paste were hung within the canopy of apple trees in Alnarp (Lomma, Sweden) at a height of approximately 1.7 m from the ground in the beginning of May 2016. The releasing devices were retrieved from the tree at 1, 7, 14, 21, and 28 days after field exposure (5 devices per date). After collection, they were placed into a 43 mL plastic vial and covered with Parafilm® for headspace collection. After 60 min of stabilization, a solid-phase microextraction (SPME) fiber (Divinylbenzene/Carboxen/Polydimethylsiloxane, Sigma-Aldrich, St Louis, MO, USA) was inserted through the Parafilm for volatile collection. After 15 min, the SPME fiber was retracted and directly inserted into a gas-chromatograph (GC) injector connected to a mass spectrometer (GC-MS 5977A MSD, Agilent technologies, Santa Clara, CA, USA) equipped with a polyethylene glycol column (DB-Wax, length 60 m, diameter 0.25 mm, df 0.25 µm, Agilent Technologies, Santa Clara, CA, USA) for 40 sec. The GC-MS started at 100 ◦C for 2 min and increased 15 ◦C each 2 min until 200 ◦C and then finished at 250 ◦C for 2 min. Prior to each volatile collection, the SPME device was conditioned in a GC port at 250 ◦C for 10 min. After headspace collection, each lure was wrapped in aluminum foil and stored at −80 ◦C until use in the subsequent trapping experiment (see below). Volatiles were identified by comparing their spectra with those published in the reference library NIST05. In addition, AA, MS, and PAA were identified by comparing their retention index with those of synthetic standards. To estimate the release rate of the main components, microcapillary tubes were partially filled with one of the compounds and placed in the same 43 mL vials described previously [24]. Sampling followed the same protocol as described above. The microcapillary tubes were weighted before and after headspace collection. The weight loss of the microcapillary tubes was then related to the area of the corresponding GC-MS peak to calibrate the amounts released. Of the three main compounds, only AA evaporated in measurable amounts using this protocol, and was therefore used for calibration.

2.3. Attraction Longevity To determine the device’s attraction longevity in the field, aged lures previously collected from the orchard and analyzed were removed from the freezer and placed in McPhail traps (Sanidad Agrícola Econex S.L., Murcia, Spain). Traps were subsequently hung in five apple orchards from the 5th to the 19th of August 2016. Orchards were located in Kivik (Skåne county, Sweden. 55◦410 N, 14◦130 E). Three orchards were organically certified and two were under integrated protection (IP). The orchards (minimum 5 ha surface) were situated a minimum of 1 km apart. In each orchard, 12 McPhail traps were placed in two different circles (diameter 14 m). Each circle was comprised of six traps loaded Insects 2019, 10, 6 4 of 15 with either PE bag or paste aged at 1, 7, 14, 21, to 28 days and with a blank trap. The two circles were 30 m apart and at least 10 m from the orchard border. Traps were hung at a height of 1.6 m and 7 m apart. Trap position was randomized within each circle at the start of the experiment. Each trap was inspected twice a week over a two-week period. In order to avoid positional effects, traps were rotated two steps in a clockwise direction within each circle at each inspection (Figure S1). The collected specimens were stored in ethanol (70 vol. %) for species identification and sex determination.

2.4. Measurement of Biological Control A field experiment was conducted in spring barley fields (H. vulgare, cv. Helium) in Ås, Norway (59◦670 N, 10◦770 E) in June and July 2016. Either a PE bag or a paste formulation was installed at the center of 25 m2 plots (N = 12). Distances between plots were at least 5 m. Three different types of lure applications were tested: (1) PE bag dispenser (300 mg total load) at vegetation height; (2) 1 mL paste-droplet with 300 mg total load on the plant (paste_1x); and (3) 3 mL paste-droplet with 900 mg total load on the plant (paste_3x). Four plots for each of the three formulation types were arranged randomly within the crop. The PE bag dispensers were hung on wooden sticks at approximately vegetation height. The wire enabled weekly adjustment to the height of the dispenser to mirror the vegetation height of the growing barley plants. The paste droplets (1 or 3 mL/plant) were applied with a 100 mL-syringe on leaves in the upper third of the barley plants. At a distance of 400 m, four control plots (25 m2) without treatment were installed with at least 5 m between the plots, as mentioned above. The 400 m distance was used to reduce the influence from either of the treatments on the control plots, as range of effect is unknown for these formulations. In the middle of each plot, a marker point (wooden stick, 60 cm) was installed. In each plot, visual inspection of lacewings (eggs and larvae of C. carnea s.l.), aphids (nymphs and adults of S. avenae and R. padi), and other natural enemies (Coccinellidae larvae, Syrphidae larvae, and parasitized aphid mummies) was performed in five differentiated sectors. Sectors were established as dispenser or marker points (=Centre; C), and 30 cm distances in the directions north (N), south (S), west (W), and east (E) of the dispenser or marker points. The observations were performed on the three plants nearest to the five marked points (C, N, S, W, E). Counted lacewing eggs were marked with a small dot on the leaf to avoid repeated counts. At the phenological stage 13 (leaf development, 3 leaves unfolded, 1st of June) of barley plants, the sectors were checked for aphids, lacewings, and other natural enemies (first record). Then the different lure types and marker points were placed in the experimental barley fields. Over an experimental period of eight weeks, the sectors were checked weekly for lacewings, aphids, and other natural enemies, as described above. Dispensers and droplets were replaced once after four weeks. Our observations were carried out within a landscape with wild inter-field vegetation, where lacewings may have had access to floral resources, overwintering sites, and alternative prey.

2.5. Data Analysis Statistical analyses were conducted with RStudio v. 1.0.143 and R software v. 3.3.3 [25], with packages lme4 v. 1.1-12 [26] and LSmeans v. 2.27-61 [27]. Fixed factors were checked for significance with the Wald-test from the car package [28]. The analysis of the release rates was carried out with linear regression models (LRM), including formulation (PE bag, Paste) and age (1, 7, 14, 21, 28 days) and their interaction as fixed factors. LRM residuals were examined visually with a QQ-plot to ascertain the normality assumption. The model for AA did not present normally distributed residuals and homogeneity of variance and was analyzed with a generalized least squares model (GLS). The amount of AA released was square root transformed and contrast was set to sum-to-zero. Tukey’s test was used for post-hoc testing of the factors age and formulation. Cumulated capture of lacewings in the McPhail traps were analyzed with generalized linear mixed models (GLMMs) with a negative binomial distribution. Initially, several exploratory models were built. The models with best fit were used to test for the possible effect of trap position within the plot Insects 2019, 10, 6 5 of 15 with each of the formulations analyzed separately as response variables. Both models included position as a fixed factor and orchard as a random effect. Additionally, to test for the possible attractiveness of unbaited McPhail traps (blanks), GLMMs were used again for both formulations separately, using catches as response variables, age or blank (1, 7, 14, 21, 28, and blank) as a fixed factor, and orchard as a random effect. Dunnett’s test was then used for post-hoc testing of the different ages against the blank. Finally, a GLMM was fitted to test the effect of lures on lacewing catches. The model included catches with both formulations as response variables, formulation, age, and their interaction as fixed factors, and orchard as a random effect. A Tukey’s test was used for pairwise multiple comparisons. In addition, two models (with each formulation analyzed separately) were fitted to test for a possible divergence from a 1:1 sex ratio in trap catches. These GLMMs included a binomial error distribution with a logit-link, age as fixed factor, and orchard as random effect. In addition to lacewings, non-parasitic wasps (Hymenoptera: Vespidae) were caught in the McPhail traps. The Fisher’s exact test was used with binomial transformed data to determine the effect of formulation (PE bag or Paste versus Blank) and the difference between them (PE bag versus Paste). The data collected in cereal included the number of lacewing eggs and larvae, aphids, and other natural enemies, and were analyzed with GLMMs with a Poisson or a negative binomial distribution and a log-link. The count of each of these groups was used as the response variables with treatment (formulation: PE bag, Paste_x1, Paste_x3, Control), sector, interaction between treatment and sector, and date as fixed factors. Additionally, sector nested within plot was included as a normally distributed random effect. The choice of the most fitting distribution was based on a test for overdispersion. After establishing the significance of the fixed factors, Tukey’s tests were carried out for pairwise comparisons between levels of each factor when necessary. All GLMMs were followed by Wald tests for statistical inference and were validated by inspecting visually the studentized residuals against the fitted values.

3. Results

3.1. Measurement of Volatile Release The emission of the loaded compounds decreased over time in both formulations (Figure1A, Table1 and Table S1). Release rates of AA, MS, and PAA were higher from the reference dispenser over the four-week period, except for PAA at day 1 (Figure1A). Although MS emission was high over the entire period of ﬁeld exposure and PAA showed an intermediate release rate, AA emission quickly decreased. Estimated release rates calibrated using AA were in the range of 0 to 1.5 mg/day. Beside the three main components, both formulations emitted lower quantities of 2-phenylethanol, benzaldehyde, benzyl acetate, and benzyl alcohol, whereas 2-heptenal and hexanoic acid were released exclusively by the paste (Figure1B,C). Climatic data were recorded every 15 min by the Lantmet weather station at Alnarp. The temperature at 1.5 m height from the soil was 18.5 ± 0.1 and 9.9 ± 0.1 ◦C, during day and night, respectively. The wind speed was 4.0 ± 0.0 and 2.4 ± 0.0 m/s, during day and night, respectively. The rainfall during the exposure time of the formulations in the ﬁeld was 17.4 mm. Insects 2019, 10, 6 6 of 15 Insects 2018, 9, x FOR PEER REVIEW 6 of 15

(A ) M ethyl salicylate Phenylacetaldehyde A cetic acid

a a a a a PE bag 6 × 109 a a Paste 1.35 (mg) AA release Mean b × 9 a ab 4 10 ab ab ab a* a b a bc 0.19 2 × 109 b* b c c b* bc* c* c* 0.02 0 cd* c* c d* d* d*

(B ) 2Heptenal− 2− Phenylethanol H exanoic acid

8 × 107

× 7 6 10 b a 4 × 107 ab ab bc bc a × 7 ab ab 2 10 c ab a ac c b b* b* a a

Mean release (±SE) 0

(C ) Benzaldehyde B enzyl acetate Benzyl alcohol

c bc bc 3 × 109 ab a 2 × 109

a* b b b b 1 × 109 a* a b* ab b b* a ab ab a 0 c* a* ab* a* a* a* a* a* bc* c*

1 7 14 21 28 1 7 14 21 28 1 7 14 21 28 Age (Days) FigureFigure 1. 1.Mean Mean release rate rate (±SE) (±SE) of ( ofA) (methylA) methyl salicylate salicylate (MS), (MS),phenylacetaldehyde phenylacetaldehyde (PAA), and (PAA), acetic and aceticacid acid(AA) (AA) from fromtwo different two different emitting emitting devices devicesat five different at five ages different of field ages exposure of field (N exposure = 5). The semi- (N = 5). Thequantitative semi-quantitative release of release AA is ofshown AA ison shown the right on y-axis. the right Letters y-axis. above Letters points above indicate points significant indicate significantdifferences differences between ages between of the ages same of formulation the same formulation and * significance and * significance between formulations between formulations at a given atage a given (LRM age and (LRM GLS, Tukey’s and GLS, test, Tukey’s p < 0.05). test, (B) Meanp < 0.05). release (B rate) Mean (±SE) release of 2-heptenal, rate (± 2-phenylethanol,SE) of 2-heptenal, 2-phenylethanol,and hexanoic acid. and ( hexanoicC) Mean release acid. (C rate) Mean (±SE) release of benzaldehyde rate (±SE), ofbenzyl benzaldehyde, acetate, and benzyl benzyl acetate, alcohol. and benzyl2-heptenal alcohol. and 2-heptenal hexanoic acid and we hexanoicre exclusively acid were found exclusively in Paste. found in Paste.

3.2.3.2. Attraction Attraction Longevity Longevity AA total total of of 348 348 adult adultC. C. carneacarnea s.l. werewere caught caught inin traps traps in inthe the attraction attraction experiment. experiment. There There was no was noeffect effect of of trap trap position position on on catch catch (Table (Table 1).1). Blan Blankk traps traps caught caught 4 4 lacewings. lacewings. All All treatments treatments were were significantlysignificantly different different against against the the blank, blank, except except for thefor 28-day-oldthe 28-day-old paste. paste. Although Although the referencethe reference device attracteddevice attracted in total a in higher total a number higher number of lacewings of lacewi thanngs the than new the paste new (Tablepaste (Table1), the 1), decline the decline over timeover in attractiontime in attraction was comparable was comparable for most offor the most dates of the (Figure dates2 ).(Figure Both formulations2). Both formulations were equally were attractiveequally duringattractive the two during first the weeks two andfirst caught weeks theand highest caught numberthe highest of catchesnumber in of the catc newesthes in lure.the newest After 21lure. days After 21 days only the reference dispenser remained attractive (Figure 2). Males and females were only the reference dispenser remained attractive (Figure2). Males and females were equally attracted equally attracted by both dispenser types independently of age, except 1-day-old paste, which by both dispenser types independently of age, except 1-day-old paste, which attracted significantly attracted significantly more females (Figure 2, Table 1). In addition to lacewings, a total of 89 wasps more females (Figure2, Table1). In addition to lacewings, a total of 89 wasps (Hymenoptera: Vespidae) (Hymenoptera: Vespidae) were caught as by-catches. Although both formulations caught a higher were caught as by-catches. Although both formulations caught a higher number of wasps than blank number of wasps than blank traps, no difference between formulations was detected (Table 1). traps, no difference between formulations was detected (Table1). Insects 2018, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/insects Insects 2019, 10, 6 7 of 15

Table 1. Results of statistical analyses.

Statistical Test Experiment Model Fixed Factor/s χ2./F df. p-Value (Distribution) Apple (volatile release) Methyl salicylate LRM (Gaussian) Formulation 967.2 1 <0.001 Age 59.9 4 <0.001 Formulation:Age 18.1 4 <0.001 Phenylacetaldehyde LRM (Gaussian) Formulation 120.6 1 <0.001 Age 40.0 4 <0.001 Formulation:Age 17.6 4 <0.001 Acetic acid GLS (Gaussian) Formulation 24.8 1 <0.001 Age 678.9 4 <0.001 Formulation:Age 2.2 4 0.7 2-heptenal LRM (Gaussian) Age 17.9 4 <0.001 2-phenyletahnol LRM (Gaussian) Formulation 16.8 1 <0.001 Age 0.6 4 0.641 Formulation:Age 8.1 4 <0.001 Benzaldehyde LRM (Gaussian) Formulation 820.7 1 <0.001 Age 17.3 4 <0.001 Formulation:Age 60.4 4 <0.001 Benzyl acetate LRM (Gaussian) Formulation 226.4 1 <0.001 Age 0.3 4 0.906 Formulation:Age 4.3 4 <0.001 Benzyl alcohol LRM (Gaussian) Formulation 316.6 1 <0.001 Age 7.9 4 <0.001 Formulation:Age 18.0 4 <0.001 Hexanoic acid LRM (Gaussian) Age 18.2 4 <0.001 GLMM (Negative Apple (trap catches) Trap position Position 2.4 5 0.795 binomial) Formulation 0.3 1 0.602 Positoin:Formulation 2.5 5 0.777 GLMM (Negative Age vs. Blank (PE bag) Age 23.2 5 <0.001 binomial) GLMM (Negative Age vs. Blank (Paste) Age 30.6 5 <0.001 binomial) GLMM (Negative Lacewing catches Formulation 12.2 1 <0.001 binomial) Age 20.0 4 <0.001 Formulation:Age 8.3 4 0.082 Sex ratio between GLMM (Binomial) Formulation 2.5 1 0.109 formulations Age 2.5 4 0.645 Formulation:Age 6.2 4 0.187 Sex ratio age (Paste) GLMM (Binomial) Age 7.1 4 0.133 Wasp (PE bag vs. Blank) Fisher’s exact test Treatment 9.2 - 0.007 Wasp (Paste vs. Blank) Fisher’s exact test Treatment 10.9 - 0.003 Wasp (PE bag vs. Paste) Fisher’s exact test Treatment 1.2 - 0.377 Barley Lacewing eggs (GLMM) Poisson Treatment 137.4 3 <0.001 Sector 574.4 4 <0.001 Date 96.5 1 <0.001 Treatment:Sector 14.3 12 0.279

GLMM (Negative Lacewing larvae Treatment 408.9 3 <0.001 binomial) Sector 44.1 4 <0.001 Date 759.7 1 <0.001 Treatment:Sector 11.4 12 0.499

Hoverﬂy larvae (GLMM) Poisson Treatment 2.1 3 0.989 Sector 2.2 4 0.994 Date 19.0 1 <0.001 Treatment:Sector 1.6 12 0.999

Ladybird (GLMM) Poisson Treatment 5.6 3 0.133 Sector 5.6 4 0.429 Date 67.7 1 <0.001 Treatment:Sector 12.0 12 0.444

Parasitoid mummies (GLMM) Poisson Treatment 30.1 3 <0.001 Sector 7.7 4 0.171 Date 50.4 1 <0.001 Treatment:Sector 9.1 12 0.697 GLMM (Negative S. avenae Treatment 628.6 3 <0.001 binomial) Insects 2019, 10, 6 8 of 15

Insects 2018, 9, x FOR PEER REVIEW 8 of 15 Table 1. Cont.

StatisticalGLMM (Negative Test Experiment S. Model avenae FixedTreatment Factor/s χ628.62./F df.3 p-Value<0.001 (Distribution)binomial) Sector 3.8 4 0.441 Sector 3.8 4 0.441 Date 455.8 1 <0.001 Date 455.8 1 <0.001 Treatment:Sector 17.5 12 0.130 Treatment:Sector 17.5 12 0.130

GLMMGLMM (Negative (Negative R. padi TreatmentTreatment 120.0120.0 33 <0.001<0.001 binomial)binomial) SectorSector 2.62.6 44 0.6310.631 DateDate 13.013.0 11 <0.001<0.001 Treatment:SectorTreatment:Sector 6.06.0 1212 0.9140.914

FigureFigure 2. 2.Mean Mean catches catches (± (±SE)SE) of of lacewing lacewing adults adults in in apple appl orchardse orchards by by two two different different emitting emitting devices devices at 1,at 7, 1, 14, 7, 21,14, and21, and 28 day28 day of fieldof fiel exposured exposure (N (N = 5).= 5). The The lower lower part part of of the the bar bar indicates indicates the the proportion proportion ofof males, males, middle middle the the proportion proportion of of females, females, and and top top the the proportion proportion of of individuals individuals that that could could not not be be sexed.sexed. Different Different uppercase uppercase letters letters above above the the bar bar indicates indicates a a significant significant difference difference of of catches catches between between thethe two two formulations formulations at at a a given given age age (GLMM, (GLMM, Tukey’s Tukey’s test, test,p p< < 0.05). 0.05). Different Different lowercase lowercase letters letters indicate indicate significantsignificant differences differences betweenbetween agesages within the the same same formulation formulation (GLMM, (GLMM, Tukey’s Tukey’s test, test, p

3.3.3.3. Measurement Measurement of of Biological Biological Control Control AA total total of of 983 983 lacewing lacewing eggs eggs and and 1965 1965 larvae larvae were were recorded recorded in the in barley the barley ﬁelds. fields. The number The number of eggs of andeggs larvae and changedlarvae changed over time over and time differed and between differed treatments between (Figuretreatments3A, Table (Figure1). The 3A, abundance Table 1). ofThe immatureabundance lacewing of immature stages variedlacewing between stages sectorsvaried (Figurebetween4A, sectors Table1 ).(Figure The most 4A, lacewingTable 1). eggs The andmost larvaelacewing were eggs recorded and larvae with thewere reference recorded formulation, with the reference followed formulation, by paste_1x followed and paste_3x. by paste_1x Very low and numberspaste_3x. were Very found low numbers in control were plots found (Figure in4 controlA). Eggs plots and (Figure larvae clustered 4A). Eggs close and tolarvae the lure clustered (sector close C), whileto the in lure the control(sector plotC), while they appeared in the control evenly plot distributed they appeared in all sectors. evenly However, distributed this in difference all sectors. is onlyHowever, statistically this difference supported is for only eggs statistically in treatments su withpported the PEfor bageggs formulation in treatments and paste_1x,with the exceptPE bag forformulation the north sectorand paste_1x, of paste_1x, except and forfor larvaethe north in the sector same treatmentsof paste_1x, (Figure and 4forA). larvae in the same treatments (Figure 4A).

Insects 2018, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/insects Insects 2018, 9, x FOR PEER REVIEW 9 of 15

Insects 2019, 10, 6 9 of 15 Insects 2018, 9, x FOR PEER REVIEW 9 of 15

FigureFigure 3. 3.Mean Mean number number (± (±SE)SE) ofof ((AA)) lacewing,lacewing, ( B(B)) other other natural natural enemies, enemies, and and ( C(C)) aphids aphids observed observed on on Figure 3. Mean number (±SE) of (A) lacewing, (B) other natural enemies, and (C) aphids observed on three barley plants during the experiment. three barleythree barley plants plants during during the the experiment. experiment.

OtherOtherOther naturalnatural natural enemies enemies were were were also also observed observed observed in the in field, thefield, field, where where wherewe we encountered encountered we encountered a total a total of 95 aof adulttotal 95 adult of 95ladybirds adultladybirds ladybirds (Coccinellidae: (Coccinellidae: (Coccinellidae: Coccinella CoccinellaCoccinella septempunctataseptempunctata septempunctata L.),L.), 2222L.), hoverfly hoverfly 22 hoverfly larvae larvae larvae(Syrphidae), (Syrphidae), (Syrphidae), and and72 and 72 72parasitized parasitizedparasitized aphid aphid aphid mummies. mummies. mummies. These These These natural natural natural enemiesenemies enemies occurredoccurred occurred in in lower lower in lowernumbers numbers numbers than than lacewing than lacewing lacewing eggs eggs eggsand andlarvaeand larvaelarvae in all inin alltreatmentsall treatmentstreatments (Figure (Figure (Figure 4A,B). 4A,B).4A,B). LadybirdLadybird Ladybird adultsadults adults and and hoverfly hoverflyhoverfly larvae larvaelarvae were werewere recorded recorded recorded in in in similarsimilarsimilar numbers numbers numbers in in all allin treatments alltreatments treatments and and and in in thein the the control, contcontrol, whereas whereaswhereas aphid aphid aphid mummies mummies mummies were were more more more abundant abundant abundant in thein controlthein control the thancontrol than in than the in restthein the rest of rest the of of treatmentsthe the treatments treatments (Figure (Figure4B). 4B). 4B).

Insects 2018, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/insects

Insects 2018, 9, x; doi: FOR PEER REVIEW Figure 4. Cont. www.mdpi.com/journal/insects Insects 2019, 10, 6 10 of 15 Insects 2018, 9, x FOR PEER REVIEW 10 of 15

Figure 4. Cumulated number of individuals (±SE) of natural enemies and aphids measured on three barleyFigure plants 4. Cumulated in the central number (C) sector of individuals of the plot (±SE) and onof na plantstural located enemies at 30and cm aphids north measured (N), south on (S), three east (E),barley and plants west (W)in the from central the central(C) sector sector. of the Different plot and letters on plants above located bars indicate at 30 cm signiﬁcant north (N), differences south (S), betweeneast (E), formulationsand west (W) in eachfrom sector the forcentral (A) lacewingsector. Different eggs (lowercase) letters above and larvae bars (uppercase)indicate significant (GLMM, Tukey’sdifferences test, betweenp < 0.050), formulations (B) ladybird in (lowercase), each sector hoverﬂy for (A) larvae lacewing (uppercase), eggs (lowercase) parasitoid and mummies larvae (bold(uppercase) lowercase) (GLMM, (GLMM, Tukey’s Tukey’s test, test,p < 0.050),p < 0.050), (B) ladybird (C) S. avenae (lowercase),(uppercase) hoverfly and R.larvae padi (uppercase),(lowercase) (GLMM,parasitoid Tukey’s mummies test, (boldp < 0.050). lowercase) (GLMM, Tukey’s test, p < 0.050), (C) S. avenae (uppercase) and R. padi (lowercase) (GLMM, Tukey’s test, p < 0.050).

Insects 2018, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/insects Insects 2019, 10, 6 11 of 15

Of the two different species of aphids (S. avenae and R. padi), S. avenae was the most abundant (Figures3C and4C). The abundance of both aphid species changed over time and differed between treatments. Both aphid species were more abundant in the control as compared to any of the treatment plots, irrespective of sector (Figure4C). S. avenae was less abundant in the paste_1x than paste_3x, whereas R. padi abundance was independent of dollop density. Lacewings were the first natural enemies recorded in the season (Figure3A). Their eggs were observed in the second week of the field trial, followed by larvae in the third week. Other natural enemies, primarily ladybird larvae, were observed in week six (Figure3B). Aphids were first detected in the second week in similar amounts in all treatments (Figure3C).

4. Discussion Adult attraction of C. carnea from a distance as well as reduction of aphid population were triggered by a blend of an induced compound and two food-signalling kairomonal cues in our field experiments. Previous studies showed that the blend used here attracted lacewings, particularly of the Chrysoperla species complex [29], and stimulated their oviposition around the formulation in cherry, apricot, and other tree species [20,22]. In the present study, we further characterized the time range of attraction in apple orchards as well as the spatio-temporal effect on lacewing density and biological control of aphids in barley. Both factors are critical in the evaluation of such formulation for a sustainable control of aphids. Together with the fact that the formulation is biodegradable and amenable for mechanical distribution, makes the paste promising for practical application, as reported earlier for similar products [23,30]. The replacement of current aphicides with this novel formulation can support ecological intensification of integrated pest management programs. The novel biodegradable formulation at the tested dosages triggered significant oviposition and larval presence in comparison to the control. While both parameters were inversely correlated with distance from the formulation, aphid population reduction remained elevated on the neighboring plants, with a possibly longer range effect. The increased presence of C. carnea eggs and larvae was most likely causing the significant reduction of both R. padi and S. avenae. However, we cannot completely rule out a possible repellence of aphid by the volatiles released from the tested formulation. For example, cis-jamone and MS were reported to repel aphid alates, including R. padi and S. avenae [31,32]. In our conditions, a lower number of lacewing eggs and larvae were measured for the higher dollop dose, possibly due to an initial repellent effect. Repellence of lacewings and other natural enemies to high loads of MS has been reported in other studies [33], along with laboratory trials showing oviposition avoidance in the presence of eggs previously laid by conspecific individuals [34]. While this last phenomenon was not observed in the present study, an increase of larval abundance over time was observed. Whether or not the tested formulations are capable of recruiting lacewing larvae from adjacent plants, however, remains unclear. The novel lacewing formulation released the three loaded volatiles over the tested period. Although the emission from the paste was lower than from the PE bag, and a lower number of lacewings were attracted to it, reduction of aphids was similar. The profile of the additionally released compounds was, however, different. These compounds are most likely breakdown products or impurities derived from synthesis, and are present in lower concentration than the main three components. How these impurities have contributed to differential attraction between the formulations is not clear. Whereas 2-phenylethanol and benzyl alcohol triggers an antennal response together with PAA in both male and female lacewings, benzyl acetate and benzaldehyde responses were not different from water [20]. Lacewing response in the field to 2-phenylethanol has been shown also to be lower than to a blend of MS, PAA, and AA [35]. Adding 2-phenylethanol to the ternary blend did not increase attraction [20]. Benzaldehyde attracted Chrysoperla plorabunda (Fitch) with captures, varying with field location [35]. No information is available on the behavioral effect of the other compounds. Insects 2019, 10, 6 12 of 15

Although lacewings were the predominant beneficial insect present in the barley field, ladybirds, hoverflies, and parasitized aphid mummies were also recorded. All ladybird adults observed belonged to the aphidophagous C. septempunctata. Therefore, although the larvae have not been identified at the species level, it seems reasonable to estimate that all of them belonged to the same species, and not to fungus feeders or pollen feeders. As lacewings arrived to the field 1–2 weeks earlier than ladybirds and approximately 6 weeks earlier than hoverfly larvae, we suspect that this early arrival was the factor shaping the composition of natural enemies in the experiment. In addition, the presence of lacewing larvae with the artificially added volatiles may have caused aversion to other predators with a reduction of oviposition due to the associated risk of intraguild predation. The semiochemical-mediated specie-specific avoidance of oviposition to conspecific or heterospecific larvae is known for several species of ladybirds and lacewings [34,36]. On this note, the higher amount of aphid mummies in the control could be due to repellence of parasitoids, as in other studies with synthetic volatiles [37], or predation of parasitized aphids by lacewing larvae. In a laboratory study, C. carnea and C. septempunctata consumed higher numbers of parasitized than un-parasitized aphids [38]. We remark here that early predation, as promoted by the lacewing formulations, is a key factor in suppression of aphid population, avoiding higher densities and potentially associated damages, as already reported for aphids [39]. While applications of HIPVs are promising for pest control, several factors need to be considered before full field transition. For example, attraction of natural enemies to synthetic kairomones in the absence of the associated herbivore prey may lead to their starvation with unpredictable consequences on the ecosystems [11]. This may be especially relevant for lacewing larvae. Further studies should consider directly rewarding recruited adults using (e.g., sugar or protein laced formulations or flower strips within or around the crop). Another issue to consider is the possible attraction of insects belonging to the fourth trophic level, which can reduce the level of herbivore predation in the long term by the targeted beneficials [40]. In order to reduce the negative effect of possible cannibalism among predators due to a low herbivore population level, HIPVs can be combined with non-crop vegetation such as flower strips, providing alternative resources (prey, nectar, and pollen) [41,42]. In our field setting, synthetic volatiles triggered oviposition of lacewings and increased lacewing larvae presence without the introduction of any reward component. Because in our experimental set-up, reduction of aphid infestation was achieved via natural enemy recruitment, we confirmed the possibility of using this method as a step towards a future ecological intensification of cultivated systems. We need to remark that our observations were collected within a landscape with wild inter-field vegetation, where lacewings may have had access to floral resources, overwintering sites, and alternative prey. It thus remains to be tested whether our result can be repeated within a larger and more intense monoculture system. An additional challenge of large-scale deployment of predator attractants is that this approach is drawing on a limited pool of predators across the landscape. While a higher density of predators is likely to occur in the field where attractants are deployed, a dilution effect may take place in neighboring fields where the same predators are pulled from. Accordingly, significant landscape- and population-level questions need to be addressed before this technology can be effectively and ethically utilized on a large scale.

5. Conclusions Although our results show a promising lacewing attraction in both a perennial and an annual cropping system, a measurement of high biological control by lacewing larvae could be observed only in barley. The distribution of aphids in apples is far more uneven and unpredictable than in an annual crop, due to characteristics related to the re-immigration from the secondary host to the overwintering host, as in the case of the rosy apple aphid. Additional experiments are thus recommended to evaluate the potential of the new formulation for biological control of aphids in apples. Insects 2019, 10, 6 13 of 15

The development of semiochemical-based methods, such as the one tested here, can offer innovative and efﬁcient approaches to control pest populations via a selective manipulation of the behavior of beneﬁcial insects. Additional research should support practical integrated pest management guidelines, including threshold value for pests and natural enemies, mechanization of application technique, and the evaluation of local non-crop vegetation as a natural enemy reservoir.

Supplementary Materials: The following are available online at http://www.mdpi.com/2075-4450/10/1/6/s1, Figure S1: Schematic representation of the trapping experiment with McPhail traps in apple orchards and how they were rotated, Table S1: Coefficient estimation for volatile release. Author Contributions: All authors contributed in planning the study. M.T., M.P., and G.T. wrote the grant proposal. J.P., G.T., and M.T. collected the field data. J.P. and T.D. carried out chemical analyses. J.P. and M.P. performed the statistical analyses. J.P. and M.T. led the writing. All authors participated in writing and improving the manuscript. Funding: This study was funded by the Swedish Research Council for Sustainable Development (Formas, Grant 2013-934), by the Research Council of Norway through the project SMARTCROP (project number 244526), and by the framework of the ERA-Net C-IPM, with funding from Formas (FP7-KBBE; project API-tree). Acknowledgments: We are thankful to the farmers who hosted our field experiments, both in Sweden and Norway, and to ISCA Tech (Riverside, USA) and Bioinnovate AB (Lund, Sweden) for providing the biodegradable prototype product. Judith Lienenleuke and Mailys Ocaña are acknowledged for their help as a part of their practical research project during field and laboratory experiments and Ian Dublon (SLU, Alnarp) for the language revision of the manuscript. Four anonymous reviewers are acknowledged for their constructive suggestions. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Ripple, W.J.; Wolf, C.; Newsome, T.M.; Galetti, M.; Alamgir, M.; Crist, E.; Mahmoud, M.I.; Laurance, W.F. World Scientists’ Warning to Humanity: A Second Notice. Bioscience 2017, 67, 1026–1028. [CrossRef] 2. Geiger, F.; Bengtsson, J.; Berendse, F.; Weisser, W.W.; Emmerson, M.; Morales, M.B.; Ceryngier, P.; Liira, J.; Tscharntke, T.; Winqvist, C.; et al. Persistent negative effects of pesticides on biodiversity and biological control potential on European farmland. Basic Appl. Ecol. 2010, 11, 97–105. [CrossRef] 3. Tittonell, P. Ecological intensification of agriculture-sustainable by nature. Curr. Opin. Environ. Sustain. 2014, 8, 53–61. [CrossRef] 4. Letourneau, D.K.; Armbrecht, I.; Rivera, B.S.; Montoya, J.; Carmona, E.J.; Daza, M.C.; Escobar, S.; Galindo, V.; Gutiérrez, C.; López, S.D.; et al. Does plant diversity benefit agroecosystems? A synthetic review. Ecol. Appl. 2015, 21, 9–21. [CrossRef] 5. Scarano, F.R. Ecosystem-based adaptation to climate change: Concept, scalability and a role for conservation science. Perspect. Ecol. Conserv. 2017, 15, 65–73. [CrossRef] 6. Batáry, P.; Dicks, L.V.; Kleijn, D.; Sutherland, W.J. The role of agri-environment schemes in conservation and environmental management. Conserv. Biol. 2015, 29, 1006–1016. [CrossRef][PubMed] 7. Rowen, E.; Kaplan, I.; Rowen, E. Eco-evolutionary factors drive induced plant volatiles: A meta-analysis. New Phytol. 2016, 210, 284–294. [CrossRef] 8. Ninkovic, V.; Markovic, D.; Dahlin, I. Decoding neighbour volatiles in preparation for future competition and implications for tritrophic interactions. Perspect. Plant Ecol. Evol. Syst. 2016, 23, 11–17. [CrossRef] 9. Heil, M.; Ton, J. Long-distance signalling in plant defence. Trends Plant Sci. 2008, 13, 264–272. [CrossRef] 10. Braasch, J.; Kaplan, I. Over what distance are plant volatiles bioactive? Estimating the spatial dimensions of attraction in an arthropod assemblage. Entomol. Exp. Appl. 2012, 145, 115–123. [CrossRef] 11. Turlings, T.C.J.; Erb, M. Tritrophic interactions mediated by herbivore-induced plant volatiles: mechanisms, ecological relevance, and application potential. Annu. Rev. Entomol. 2018, 63, 433–452. [CrossRef][PubMed] 12. Peñaflor, M.F.G.V.; Bento, J.M.S. Herbivore-Induced Plant Volatiles to Enhance Biological Control in Agriculture. Neotrop. Entomol. 2013, 42, 331–343. [CrossRef] 13. Silva, D.B.; Weldegergis, B.T.; Van Loon, J.J.A.; Bueno, V.H.P. Qualitative and Quantitative Differences in Herbivore-Induced Plant Volatile Blends from Tomato Plants Infested by Either Tuta absoluta or Bemisia tabaci. J. Chem. Ecol. 2017, 43, 53–65. [CrossRef][PubMed] Insects 2019, 10, 6 14 of 15

14. Landolt, P.J.; Alfaro, J.F. Trapping Lacanobia subjuncta, Xestia c-nigrum, and Mamestra configurata (Lepidoptera: Noctuidae) with acetic acid and 3-Methyl-1-butanol in controlled release dispensers. Environ. Entomol. 2000, 30, 656–662. [CrossRef] 15. Knight, A.L.; Hilton, R.; Basoalto, E.; Stelinski, L.L. Use of glacial acetic acid to enhance bisexual monitoring of tortricid pests with kairomone lures in pome fruits. Environ. Entomol. 2014, 43, 1628–1640. [CrossRef] 16. Jones, V.P.; Horton, D.R.; Mills, N.J.; Unruh, T.R.; Baker, C.C.; Melton, T.D.; Milickzy, E.; Steffan, S.A.; Shearer, P.W.; Amarasekare, K.G. Evaluating plant volatiles for monitoring natural enemies in apple, pear and walnut orchards. Biol. Control 2016, 102, 53–65. [CrossRef] 17. El-Sayed, A.M.; Knight, A.L.; Byers, J.A.; Judd, G.J.R.; Suckling, D.M. Caterpillar-induced plant volatiles attract conspecific adults in nature. Sci. Rep. 2016, 6, 1–14. [CrossRef] 18. Gadino, A.N.; Walton, V.M.; Lee, J.C. Evaluation of methyl salicylate lures on populations of Typhlodromus pyri (Acari: Phytoseiidae) and other natural enemies in western Oregon vineyards. Biol. Control 2012, 63, 48–55. [CrossRef] 19. Aqueel, M.A.; Collins, C.M.; Raza, A.B.; Ahmad, S.; Tariq, M.; Leather, S.R. Effect of plant nutrition on aphid size, prey consumption, and life history characteristics of green lacewing. Insect Sci. 2014, 21, 74–82. [CrossRef] 20. Tóth, M.; Szentkirályi, F.; Vuts, J.; Letardi, A.; Tabilio, M.R.; Jaastad, G.; Knudsen, G.K. Optimization of a phenylacetaldehyde-based attractant for common green lacewings (Chrysoperla carnea s.l.). J. Chem. Ecol. 2009, 35, 449–458. [CrossRef] 21. Pickett, J.A.; Khan, Z.R. Plant volatile-mediated signalling and its application in agriculture: Successes and challenges. New Phytol. 2016, 212, 856–870. [CrossRef][PubMed] 22. Koczor, S.; Knudsen, G.K.; Hatleli, L.; Szentkirályi, F.; Tóth, M. Manipulation of oviposition and overwintering site choice of common green lacewings with synthetic lure (Neuroptera: Chrysopidae). J. Appl. Entomol. 2015, 139, 201–206. [CrossRef] 23. Teixeira, L.A.F.; Mason, K.; Mafra-Neto, A.; Isaacs, R. Mechanically-applied wax matrix (SPLAT-GBM) for mating disruption of grape berry moth (Lepidoptera: Tortricidae). Crop Prot. 2010, 29, 1514–1520. [CrossRef] 24. Tasin, M.; Bäckman, A.C.; Bengtsson, M.; Varela, N.; Ioriatti, C.; Witzgall, P. Wind tunnel attraction of grapevine moth females, Lobesia botrana, to natural and artificial grape odour. Chemoecology 2006, 16, 87–92. [CrossRef] 25. R Core Team R. A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2016. 26. Bates, D.; Machler, M.; Bolker, B.; Walker, S. Fitting Linear Mixed-Effects Models Using lme4. J. Stat. Softw. 2015, 67, 1–48. [CrossRef] 27. Lenth, R.V. Least-Squares Means: The R Package lsmeans. J. Stat. Softw. 2016, 69.[CrossRef] 28. Fox, J.; Weisberg, S. An {R} Companion to Applied Regression, 2nd ed.; Sage: Thousand Oaks, CA, USA, 2011. 29. Koczor, S.; Szentkirályi, F.; Birkett, M.A.; Pickett, J.A.; Voigt, E.; Tóth, M. Attraction of Chrysoperla carnea complex and Chrysopa spp. lacewings (Neuroptera: Chrysopidae) to aphid sex pheromone components and a synthetic blend of floral compounds in Hungary. Pest Manag. Sci. 2010, 66, 1374–1379. [CrossRef] 30. Schorkopf, D.L.P.; Spanoudis, C.G.; Mboera, L.E.G.; Mafra-Neto, A.; Ignell, R.; Dekker, T. Combining attractants and larvicides in biodegradable matrices for sustainable mosquito vector control. PLoS Negl. Trop. Dis. 2016, 10, e0005043. [CrossRef] 31. Pettersson, J.; Smart, L.E. Winter host component reduces coloniza- tion by bird-cherry-oat aphid, Rhopalosiphum padi (L.) (Homop-tera, Aphididae), and other aphids in cereal fields. J. Chem. Ecol. 1994, 20, 2565–2574. [CrossRef] 32. Bernasconi, M.L.; Turlings, T.C.J.; Ambrosetti, L.; Bassetti, P.; Dorn, S. Herbivore-induced emissions of maize volatiles repel the corn leaf aphid, Rhopalosiphum maidis. Entomol. Exp. Appl. 1998, 87, 133–142. [CrossRef] 33. Snoeren, T.A.L.; Mumm, R.; Poelman, E.H.; Yang, Y.; Pichersky, E.; Dicke, M. The herbivore-induced plant volatile methyl salicylate negatively affects attraction of the parasitoid Diadegma semiclausum. J. Chem. Ecol. 2010, 36, 479–489. [CrossRef][PubMed] 34. Ruzicka, Z. Further evidence of oviposition-deterring allomone in chrysopids (Neuroptera: Chrysopidae ) Larvae of a eurytopic Nearctic species Chrysopa oculata Say (Neuroptera: Chrysopi-dae) mark substrates they walk over with an oviposition-deterring pheromon. Eur. J. Entomol. 1998, 95, 35–39. Insects 2019, 10, 6 15 of 15

35. Jones, V.P.; Steffan, S.A.; Wiman, N.G.; Horton, D.R.; Miliczky, E.; Zhang, Q.H.; Baker, C.C. Evaluation of herbivore-induced plant volatiles for monitoring green lacewings in Washington apple orchards. Biol. Control 2011, 56, 98–105. [CrossRef] 36. Seagraves, M.P. Lady beetle oviposition behavior in response to the trophic environment. Biol. Control 2009, 51, 313–322. [CrossRef] 37. Yazdani, M.; Baker, G. A plant volatile-based attractant formulation is not attractive to Diadegma semiclausum (Hymenoptera: Ichneumonidae). Austral Entomol. 2018, 57, 359–364. [CrossRef] 38. Klug, T.; Meyh, R. Intraguild predation on the aphid parasitoid Lysiphlebus fabarum (Marshall) (Hymenoptera: Aphidiidae): Mortality risks and behavioral decisions made under the threats of predation. Biol. Control 2002, 25, 239–248. 39. Porcel, M.; Andersson, G.K.S.; Pålsson, J.; Tasin, M. Organic management in apple orchards: Higher impacts on biological control than on pollination. J. Appl. Ecol. 2018, 55, 2779–2789. [CrossRef] 40. Orre, G.U.S.; Wratten, S.D.; Jonsson, M.; Hale, R.J. Effects of an herbivore-induced plant volatile on arthropods from three trophic levels in brassicas. Biol. Control 2010, 53, 62–67. [CrossRef] 41. Simpson, M.; Gurr, G.M.; Simmons, A.T.; Wratten, S.D.; James, D.G.; Leeson, G.; Nicol, H.I.; Orre-Gordon, G.U.S. Attract and reward: Combining chemical ecology and habitat manipulation to enhance biological control in ﬁeld crops. J. Appl. Ecol. 2011, 48, 580–590. [CrossRef] 42. Hatt, S.; Boeraeve, F.; Artru, S.; Dufrêne, M.; Francis, F. Spatial diversiﬁcation of agroecosystems to enhance biological control and other regulating services: An agroecological perspective. Sci. Total Environ. 2018, 621, 600–611. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).